[Bug-openmcl] minor adjust-array fix, displaced-array question

[bryan o'connor]

adjust-array wasn't properly filling in the fill-pointer of the
new array when not specified but present in the old array.

this fixes the ~dozen new adjust-array test failures, except..

make-array-1 creates a displaced-array when a fill-pointer is
requested.  i imagine this is an internal optimization, but ansi
says that if :displaced-to is nil the new array is not displaced
(such that #'array-displacement => nil).

if the use of displaced-arrays is intentional, array-displacement
would need to know enough to return NIL in this case.

						...bryan

ps.  it looks like the %make-[s/d]float bug regressed in the latest
      change to acos.


Index: lib/arrays-fry.lisp
===================================================================
RCS file: /usr/local/tmpcvs/ccl-0.14-dev/ccl/lib/arrays-fry.lisp,v
retrieving revision 1.5
diff -u -r1.5 arrays-fry.lisp
--- lib/arrays-fry.lisp	18 Jan 2004 03:53:09 -0000	1.5
+++ lib/arrays-fry.lisp	15 Feb 2004 23:10:10 -0000
@@ -254,7 +254,9 @@
                                         displaced-to
                                         displaced-index-offset
                                         nil
-                                       fill-pointer
+                                       (or fill-pointer
+                                           (and 
(array-has-fill-pointer-p array)
+                                                (fill-pointer array)))
                                         initial-element 
initial-element-p
                                         initial-contents 
initial-contents-p
                                         size)))